Well that sounds just great: Scientists make sarcasm detector

The lowest form of wit is sometimes the hardest to detect, in writing at least. And depending on how subtle the delivery - or how callow or tin-eared the listener - sarcasm can sometimes be missed in conversation as well, sailing over heads like a mis-hit corner kick.

But now, in the most important communications advance since the invention of the telephone (sarcasm detected), a team of Netherlands-based scientists have come up with an emoticon-based algorithm for recognizing ironic speech.

"We extracted acoustic parameters such as pitch, speaking rate, and energy from speech, then used automatic speech recognition to transcribe the speech into text for sentiment analysis," said Xiyuan Gao of the Speech Technology Lab at the University of Gronigen, discussing the team's mind-blowing work.

The researchers "assigned emoticons to each speech segment, reflecting its emotional content," Gao said, claiming the technique "leverages the combined strengths of auditory and textual information along with emoticons for a comprehensive analysis."

Surely a science Nobel looms if the snark detector can only be tweaked and fine-tuned a bit more. And on that note, brace yourselves: The aim is to enhance the system by expanding its repertoire of "dry wit," according to Gao.

First, however, the team is to present its research to the eagerly-awaited (possible sarcasm detected) joint meeting of the Acoustical Society of America (ASA) and the Canadian Acoustical Association (CAA). A standing room-only event, if there ever was one.

Later, once the buzz around the conference fades, Gao and colleagues Shekhar Nayak and Matt Coler hope to add "a range of expressions and gestures people use to highlight sarcastic elements in speech" - elements they concede "need to be better integrated into our project."

You don't say.